The present invention relates to a method and system for assessing the risks and/or exposures associated with financial transactions using various statistical and probabilistic techniques. Specifically, the present invention relates to a method and system for identifying plausible sources of error in data used as input to financial risk assessment systems using Bayesian belief networks as a normative diagnostic tool to model relationships between and among inputs/outputs of the risk assessment system and other external factors.
Legal claims defining the scope of protection, as filed with the USPTO.
1. A computer-implemented method for identifying plausible sources of error in a risk assessment system, comprising: identifying, by a computer, a first variable and a second variable of the risk assessment system, wherein an initial distribution of the first variable is a first hypothesis and an initial distribution of the second variable is a second hypothesis; implementing, by the computer, a Bayesian network to represent implications between the first and second variables; determining, by the computer, an initial probability of the first hypothesis that the first variable has not changed given the second hypothesis that the second variable has not changed, wherein the initial probability is based on a state of knowledge at the time of determining the initial probability; receiving, by the computer, data regarding the first and second variables after determining the initial probability of the first hypothesis; identifying, by the computer, a change of value in the first or second variable; determining, by the computer, by probabilistic induction at least one cause of the change of value in the first or second variable, wherein the at least one cause is a plausible source of error; and determining, by the computer, the plausibility that the change is an error in the data by evaluating the initial probability of the first hypothesis based on the at least one cause.
2. The method of claim 1 , wherein the initial distribution of the first hypothesis is 0.95 and indicates a 95% certainty that the first hypothesis is correct.
3. The method of claim 1 , wherein the first and second hypotheses are assumed to be true.
4. The method of claims 1 , wherein determining the plausibility that the change is an error in the data comprises determining a source that is not likely causing the change.
5. The method of claim 1 , wherein the at least one variable of the risk assessment system comprises input data of the risk assessment system.
6. The method of claim 1 , wherein the at least one variable of the risk assessment system comprises output data of the risk assessment system.
7. The method of claim 1 , wherein the at least one variable of the risk assessment system comprises data external to the risk management system but related to the risk assessment system.
8. The method of claim 1 , wherein the risk assessment system comprises a pre-settlement exposure server.
9. The method of claim 1 , wherein the at least one variable of the risk assessment system comprises observable information.
10. The method of claim 1 , wherein the at least one variable of the risk assessment system comprises a plurality of variables, and wherein a first one of the plurality of variables implicates a second one of the variables.
11. The method of claim 1 , wherein determining a first hypothesis about the at least one variable comprises: hypothesizing that the at least one variable has not changed in value.
12. The method of claim 1 , wherein providing an initial probability of the first hypothesis comprises: providing a prior probability of the at least one variable; and providing an initial conditional probability of the at least one variable.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
October 27, 2000
October 18, 2011
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.